Resistant Starch

Whether or not you decide to implement a ketogenic diet, you'll want to start supplementing with resistant starch (RS). RS is a form of starch that is indigestible by humans. We don't have the enzymes required to break it down, so we rely on microbes to do that for us. It turns out that our health is much improved when we have a robust population of mostly beneficial microbes busily fermenting carbohydrates into fats in our colons. Gorillas have giant colons which allow them to eat vast amounts of cellulose to feed cellulose digesting microbes which produce fats. In comparison, human colons are too small to provide for very much cellulose digestion. Our small colons have been cited as evidence that we gave up fermenting things in preference for eating fat and meat directly. Although we apparently did learn how to get a lot more fat directly in our diets, it looks like we may have also switched from fermenting cellulose carbohydrates to fermenting more concentrated forms of carbohydrates such as roots and pollen, which don't require such a gigantic colon. RS is just such a carbohydrate.

There is a lot of research out there on RS, but very little public awareness of it. I am only aware of it thanks to "Tatertot" and Richard Nikoley who have been posting about it on Nickoley's blog. It is only a matter of time before books start coming out popularizing the use of RS for health. To summarize what I've learned so far, mostly from Nikoley's blog, here are some bullet points. I can personally vouch for some of the effects listed below.

Resistant Starch (RS)

Main Effects

• Lowers blood sugar
• Increases body temperature
• Increases mineral absorption
• Increases beneficial bacteria
• Decreases bad bacteria
• Improves colon health
• Improves elimination habits
• Improves cholesterol profile

 Other Effects

• Increases serotonin production in the gut
• Causes increase in length and vividness of dreams

 How Does RS Work

• Indigestible by humans, eaten by bacteria
• Provides safe transit for beneficial bacteria to the colon
• Preferentially feeds beneficial bacteria
• Allows beneficial bacteria to outcompete bad bacteria
• Attracts bad bacteria away from bodily tissues and transports them to the colon where they can be more easily disposed of
• In the colon, bacteria break down RS to produce butyrate and other short chain fatty acids (SCFAs)
• Butyrate is the preferred fuel of the cells which line the colon (colonocytes)
• Colonocytes can only run on butyrate when it is produced in ample amounts via fermentation in the colon
• SCFAs are healthy fats, also found in coconut oil
• Just as coconut oil is good for skin, the SCFAs produced in the colon are good for the colon
• SCFAs are highly ketogenic fats (easily broken down to produce ketones)
• Ketones are beneficial for brain health

How to Get RS

• Aim to get 4 tablespoons per day of Bob’s Red Mill Unmodified Potato Starch
• Start with 1 tablespoon per day
• Add another tablespoon per day every 3 to 7 days until you get to 4 per day
• Just mix it with plain water and drink it
• Take it in 2 or more servings throughout the day
• If it makes you too gassy just back off a bit and increase again after a few days
• It may be better to take it in between meals to avoid excessive gassiness
• The potato starch must be raw and unmodified (heat destroys RS)
• Although heat destroys raw RS, when starches such as rice are cooked and then cooled, RS re-forms during the cooling process
• Supplement with food sources of RS such as green bananas, parboiled rice that has been cooked and then cooled (reheating is ok), cooked and cooled potatoes, others…

Click here for a printable version of the bullet points above.

Resistant starch ought to be helpful in fighting cancer.

Does Eating a Ketogenic Diet Help Fight Cancer?

A cursory search reveals two very interesting journal articles.

The Ketogenic Diet Is an Effective Adjuvant to Radiation Therapy for the Treatment of Malignant Glioma (full text available)
Mohammed G. Abdelwahab et al.
May 2012

Ketogenic Diets Enhance Oxidative Stress and Radio-Chemo-Therapy Responses in Lung Cancer Xenografts
Bryan G. Allen et al.
July 2013

These are are both mouse studies which used the commercially available shake "KetoCal" to implement the ketogenic diet. KetoCal is used to implement a ketogenic diet in kids with epilepsy.  Abdelwahab et al. studied mice given brain cancer, while Allen et al. looked at mice implanted with lung cancer in their torso (ie. not in the lungs themselves).

In most of these experiments you'll see four groups.

• Controls: Mice fed the standard lab chow diet and given no cancer treatments.
• Ketogenic Only: Mice put on a ketogenic diet and given no cancer treatments.
• Radiation/Chemo Only: Mice on the standard diet given cancer treatments.
• Ketogenic with Radiation/Chemo: Mice on the ketogenic diet given cancer treatments.

The question is which group would you like to be in if you had a choice.

Grains of Salt

The Abdelwahab study was funded by the company that makes KetoCal. They state: "The funders had no role in data collection and analysis, decision to publish or preparation of the manuscript, but did input to the study design." The Allen study was funded by an NIH grant.

These are studies of rodents implanted with cancerous cells, so the results are not guaranteed to be replicated in humans who've developed cancer endogenously. I've read that successes in rodent cancer research have not always translated into similar successes in human treatments, to the frustration of the cancer research community. It is easy to keep in mind that the efficacy of the treatments in the studies when applied to humans may or may not mirror those seen in the rodents.

Aside

It would be wrong to take the fact that these are rodent studies as a license to simply write them off without consideration. The reason thousands of studies are done each year on rodents is not because humans are fascinated with rodents, or because rodents are convenient subjects, it is because the results are relevant to human biology.

Hodgkin and Huxley received a Nobel Prize for their work in the 1950's elucidating the electrical workings of neurons. The neurons they studied happened to be those of the giant squid, because squid neurons were big enough to manipulate and measure using the tools available. If there were Hodgkin and Huxley haters at that time, they probably would have dismissed the research by arguing that insights about giant squid neurons were irrelevant to human biology.

The most powerful aspect of rodent studies is that they can be setup as true experiments, in which the rodents are randomly assigned to experimental groups, the manipulation is conducted, and results measured. The results which come from such experiments are highly informative and allow the researcher to conclude that "treatment X caused result Y."

In epidemiological research, on the other hand, there is no random assignment to groups, and therefore no statements as to cause can be made based on the results. The epidemiological researcher can only say that "in this analysis, people who ate more vegetables had less heart attacks." They can't say that eating vegetables causes less heart attacks, because it could be that vegetables actually cause heart attacks but people who eat vegetables also tend not to smoke.

So, if there is a good reason to think that a particular result in rodents would not apply in humans, that's one thing. But to dismiss a result in rodents simply because it is not a human study is a bad habit. On the other hand, if someone wants to shrug off the results of an epidemiological study without a logical reason then I can't really blame them.

Further, the results discussed below are not random. The effects of no treatment vs chemotherapy and radiation are as one would predict based on the fact that chemotherapy and radiation are the current standard of care for cancer. So there is no a priori reason to think that the effect of adding a ketogenic diet to chemo/radiation therapy in humans would be radically different from the effect seen in rodents below.

Enough of that. What did these studies find?

Tumor Growth

In these tumor growth graphs, you want to be in the line with lowest slope, representing the group whose tumors grew the slowest. Allen et al. looked at several different types of radiation treatment, resulting in the multiple graphs below.

I don't know if it is really necessary but in order to avoid any possible copyright issues for the Allen paper, which is not freely available, I've created new graphs from scratch in Excel by eyeballing the originals. It allows me to change the emphasis anyway. The treatments involving ketosis are displayed in dotted lines.

Based on these results it looks like ketosis alone is of little or no help in inhibiting tumor growth, but adding ketosis to chemo and/or radiation results in the least tumor growth obtained, sometimes by a wide margin.

Adbelwahab et al. did not report changes in tumor weight.

Survival

Regardless of changes in tumor weight, we want to know which group of mice lived the longest. The survival graphs are more difficult to replicate so I'll just describe Allen et al.'s results.

Allen et al. Experiment 1
Controls: last survivors died on day 45
Radiation: last survivors died on day 50
Keto-Radiation: last survivors died on day 80
Keto-Chemo-Radiation: 20% still alive on day 80

Allen et al. Experiment 2
Controls: last survivors died around day 38
Keto: last survivors died on day 40
Radiation: Roughly 22% still alive at day 80
Keto-Radiation: 100% still alive at day 80*

Allen et al. Experiment 3
Controls: last survivors died on day 41
Keto: last survivors died on day 35
Radiation: last survivors died on around day 48
Keto-Radiation: 100% survived past day 50, but then only 20% made it past day 60*

* Important to note that the mice in experiment three were on the ketogenic diet for seven days total, whereas the mice in experiment two were on the ketogenic diet for 81 days at which point the experiment was terminated. The two experiments had different radiation regimens as well.

Abdelwahab et al. have their survival graphs split into two pieces, which I have helpfully smushed together into a single graph below. The x-axis is day number, the y-axis is percentage of mice still alive.

In the Abdelwahab et al. study, roughly 80% of the mice in the keto-radiation group survived indefinitely. Did they still have tumors? Here's how the authors describe the results.

"Nine out of the 11 animals treated with [KetoCal] in combination with radiation were apparently cured of their implanted tumor...In the animal[s] treated with radiation and [KetoCal] the presence of growing tumor can be seen for the first [few] weeks following implantation, reaching a maximum bioluminescent [around] day 9 following tumor implantation. This was followed by a near exponential decline that approaches background levels 60 days following implantation. Bioluminescence remained undetectable and on day 104 post-implantation the 9 surviving animals treated with radiation and [KetoCal] were switched from KC to the standard rodent chow. There was no detectable recurrence of tumor as demonstrated by the continued absence of detectable bioluminescent signal. The animals were sacrificed on day 299. Histological evidence ... from the apparently cured [KetoCal] plus radiation animals showed no evidence of tumor cells in or near the area of implantation."

 
In summary, the group that lived the longest in all four experiments was the one that combined radiation and/or chemo with a ketogenic diet.

Body Weight Maintenance

There is concern that since the ketogenic diet is a restricted diet it will lead to weight loss. In Allen et al. experiment 1, the authors report:

"comparisons of animal weights show that all treatments were well tolerated as shown by a lack of significant weight change"

In fact, the ketogenic mice are the heaviest group in 13 out of the 17 days displayed in Allen et al.'s graph.

Abdelwahab et al. divided their figure into two pieces again, so I have again helpfully mangled the two pieces into a single figure in which they appear on the same horizontal time scale.

There was essentially no difference in bodyweight between the controls fed the standard mice chow and the mice fed KetoCal. However, the KetoCal fed mice who were given radiation therapy retained their weight much better than those fed the standard diet. Perhaps that is to be expected given that their tumors disappeared and most of them survived.

In summary, it appears that the ketogenic diet had a neutral to positive effect on the bodyweights of the mice in these studies.

Will These Results be Replicated in Human Studies?

No one knows yet. Only a few small and flawed trials have been conducted in humans. However, clinical trials are now starting up, as previously discussed (KETOLUNG and KETOPAN). The fact that these clinical trials are in the works is very significant. It means that:

1. Enough evidence has been published already to provide a scientific justification for conducting clinical trials, which are not trivial undertakings.
2. The evidence is strong enough to convince not just a few curious oncologists, but entire panels of conservative experts at agencies such as the NIH that fund such trials, as well as the institutional review boards that approve trial protocols for human safety.
3. These trials are proceeding despite the fact that there is no drug company to push them along in hopes of discovering the next blockbuster drug (KetoCal is not required for ketosis, see below).

In other words, the idea of using a ketogenic diet to help fight cancer is not some kooky idea from the natural health fringe. It may not yet be mainstream oncology, but it is backed by reasonably compelling scientific findings.

Translating drug therapies from rodents to humans is difficult and incredibly risky. In contrast, the ketogenic diet is an established medical therapy as well as something that many people already do by choice. Allen et al. included the following sidebar titled "Translational Relevance" in their article.

"Ketogenic diets are high in fat, low in carbohydrates, and are well established as an alternative therapy for childhood epilepsy. This report shows that a ketogenic diet enhances radio-chemo-therapy responses as well as enhancing oxidative stress in human lung cancer xenografts. As ketogenic diets are an established therapy in humans, these studies may be rapidly translated into the clinical setting, potentially allowing for improved cancer control without added normal tissue toxicity."

It is lucky that a therapy offering potential benefits such as those above doesn't have to undergo many years of testing for safety before it can be offered as an option for patients.

Final Notes

Although I referred to the mice in the above studies as being on the "ketogenic diet," I should have said "the mice given the KetoCal diet." It is possible that it wasn't the ketosis that helped those mice but some component of the KetoCal shake. However, that doesn't seem all that likely. In fact, it is kind of amazing that mice with cancer could recover on a KetoCal only diet, since KetoCal is just an industrial food-like substance consisting primarily of cancer-promoting soybean oil and fortified with vitamins and minerals. It's basically a ketogenic version of Ensure. Such foods are usually associated with poor outcomes no matter how hard they try to imitate the components of real food.

So we can assume for now that it was the ketogenic-ness of the diet that produced the anti-cancer effect seen in these studies. It is encouraging to consider that these researchers might have gotten even better results with a ketogenic mouse diet consisting of healthier foods.

That's all I have on these two articles for now.

Ketosis

So in the first post we discussed how cancer cells appear to require large amounts of glucose to survive. If that is the case, the obvious question is whether or not it might be possible to attack cancer by lowering blood glucose levels. Theoretically if we could reduce blood glucose levels to zero then the cancer would perish, but of course so would we. Even if you do not eat any carbohydrates at all, your liver will make glucose out of proteins and fats in order to maintain your blood glucose at the minimum level necessary for survival.

Just because we can't lower our blood glucose levels to zero doesn't mean that striving to keep our blood glucose levels at the lower end of the normal range wouldn't be helpful for treating or preventing cancer. Blood glucose levels can vary between somewhere around 80 to over 200 of whatever the measure is. Normal fasting blood glucose is somewhere around 90. After carbohydrate-containing meals, blood glucose spikes into the mid to upper 100's. It could be that cancer cells prefer a high fasting blood glucose, or it could be that they benefit most from the frequent spikes caused by carb-heavy meals. So anyone trying to fight cancer by targeting blood sugar would want to try to lower both fasting blood glucose and avoid frequent/high blood sugar spikes.

The only real option for drastically lowering blood sugar is to eat a "ketogenic" diet. On a ketogenic diet you basically substitute fat for carbohydrates, keeping carbohydrate intake at less than 50 grams per day, without increasing protein intake. When carbs are absent from the diet and protein intake is moderate, the body switches from burning glucose as a primary fuel to burning fat. The process of burning fat produces molecules called "ketones" as a byproduct. Ketones are water soluble, so they disperse throughout the body, and mitochondria will switch to burning them instead of glucose or fats. This allows many of the cells which supposedly require glucose, particularly in the brain, to replace glucose use with ketones. Eating a ketogenic diet and producing and burning ketones for fuel is called being in "ketosis."

Ketosis is a natural state. It starts to kick in anytime we go without food for awhile, something that our ancestors probably did regularly. Even people on normal diets supposedly start transitioning to ketosis while sleeping overnight before breaking their fast in the morning or afternoon. What's really cool though is that ketosis appears to have a lot of positive affects on cellular processes, especially in the brain. In particular, ketogenic diets are becoming well-known for their success in treating childhood epilepsy.

But back to cancer. So a ketogenic diet will put a stop to post-meal blood sugar spikes, which could help with cancer. Fasting blood glucose will probably go down (although over an extended time ketogenic diets will raise fasting blood glucose, which could be bad, but we don't need to worry about that at this point). Through exercise or perhaps other tricks someone in ketosis can attain a blood sugar level so low that it would cause a hypoglycemic crisis in anyone not in ketosis, but yet feel no effects of hypoglycemia. The cellular energy crisis which causes the hypoglycemic symptoms never happens because the person in ketosis has plenty of fat and ketones available to with which to run cellular processes. But that's about as low as one can expect to get blood sugar, and only transiently.

So avoiding post-meal blood glucose spikes and lowering fasting blood sugar could potentially help fight cancer. It certainly wouldn't be expected to hurt. But ketosis is a complex physiological state and what's interesting is that it might have important effects on cancer apart from its effect on blood glucose, especially in combination with chemo and/or radiation. More as soon as I can get to it...

Vegetable Oils Increase Tumor Growth

[This post derives from a post on cancer just put up by Peter Dobromylskyj at Hyperlipid.]

I just want to make two points in this post.

Vegetable Oils Increase Tumor Growth

I'm sure that a book or two could be written exploring all of the ways in which vegetables are horrible for human health (a cursory look on Amazon didn't reveal any, which is kind of surprising). Or one could dedicate a blog just to exploring all of the research and mechanisms involved. There is a lot of information out there already if one is interested and I don't have the time or energy to pull it all up and summarize it.

However, it makes sense to illustrate at least one example of the dangers of vegetable oils. Peter's post highlights some research by Sauer et al in the 1980's which found that vegetable oils but not other fats promoted tumor growth in mice. The following graph illustrates the data in Table 3 of Sauer's paper. The dotted line represents tumor growth in the mice in the control condition (the first bar).

You can more than double the growth of these mouse tumors by providing them with linoleic and arachidonic acid. Linoleic acid is found primarily in vegetable oils. For example, soybean oil is 51% linoleic acid, whereas butter is only 2%. Arachidonic acid is found in some animal products and is also made from linoleic acid in the body. In contrast, palmitic, stearic and oleic acids did not promote tumor growth relative to the control condition. Palmitic and stearic acids are the most common saturated fats found in animal fats (stearic is also the main fat found in cocoa butter). Oleic acid is found in both animal fats and plant sources such as olive oil.

In the chart below, the blue bars represent the percentage of fatty acids that DON'T promote tumor development, and the red bars represent the percentage of fatty acids that DO promote tumor development according to the Sauer study above. Arachidonic acid percentages weren't listed, so I assume they must be small. I left out any fatty acids not included in the above study.

If you don't want to promote tumor development, it looks like a BAD idea to eat soybean oil or even supposedly healthy flax oil. Butter on the other hand looks great.

Current Trials of Ketogenic Diets for Cancer Treatment in Humans are Seriously Misguided

[Adapted from a comment I left on Hyperlipid.]

I haven't gotten to the logic of using a low-carb ketogenic diet as a cancer treatment, but that's where I've been heading. A low-carb ketogenic diet is by necessity a very high fat diet (80% fat or so usually). There are reasons to think it may be therapeutic for cancer patients. However, only a few human studies have been completed, although several clinical trials are in the works.

Unfortunately, despite the cancer-promoting effects of vegetable oils, all but one of the human trials of ketogenic diets for cancer I can get my hands on about have put patients on high vegetable oil diets. For example Schmidt et al 2011 had their participants eat two "liquid meals" a day composed of the following ingredients.

Highly fermented yoghurt-drink:

• skimmed milk
• plant oil mixture
• pectin

Vegetable oil mixture:

• line seed oil
• canola oil
• walnut oil
• MCT
• grape seed oil
• argan-oil
• pumpkin seed oil

Protein preparation:

• milk-protein

Additionally: "patients were encouraged to add additional servings (1 tablespoon each) of the oil mixture or other oils from olives, flaxseed and hempseed to the three principle meals." This is a crazy amount of vegetable oil. Of course, they got poor compliance and lots of adverse reactions. Nevertheless, they concluded the diet "might improve aspects of quality of life and blood parameters in some patients with advanced metastatic tumors."

A recent research protocol from Dr. Eugene Fine, who appears to be a thoughtful scientist, nevertheless stipulates "unsaturated oils for cooking."

Two clinical trials are starting up, aiming to investigate the use of a ketogenic diet for lung cancer (see KETOLUNG) and pancreatic cancer (see KETOPAN).

There is a good explanation of the lung cancer trial here, DEFINITELY READ IT. It explains the basic logic behind the use of ketogenic diets for cancer treatment. They will be implementing a 90% fat diet. HOWEVER, if you read the fine print in the study design you will see that they intend to use "Ketocal" shakes to implement the diet. Ketocal is a commercial product you can buy on Amazon. You can read the list of ingredients here. The first three ingredients are:

• Hydrogenated soybean oil
• Whole milk
• Refined soybean oil

Unbelievable. This is exactly what you would feed a mouse in order to give it cancer, or encourage its existing tumors to grow rapidly. It is truly amazing. If this trial were going on at M.D. Anderson I would fly down there and physically prevent you from participating in it. It is incredibly disappointing that good studies of this topic don't even seem to be in the hopper yet, so we can't expect answers for who knows how many more years.

To end on a more positive note, the one exception to the above studies used medium chain triglycerides (MCTs). MCTs are technically a vegetable oil I think, but one which is highly ketogenic and not an omega-6 oil, which is what you want to avoid. The trial was only in two pediatric patients, but the results seemed pretty positive compared to the vegetable oil trials.

Luckily you don't need to enroll in a clinical trial to implement a not insane version of a ketogenic diet yourself. At the very least, stop eating all sources of vegetable oils ASAP. Not easy when you are living in a hotel, but what can you do. Unfortunately that means no more fried calamari.

Normal vs Cancer Cell Metabolism and How PET Scans Work

 [This is my current understanding of the latest ideas on how cellular metabolism works with respect to cancer based on reading a few articles.]

Cells require ATP in order to survive, as well as fulfill whatever duties they have to the organism as a whole. ATP isn't just floating around for free in unlimited amounts for cells to take up and use, so each cell has to manufacture it's own supply. For people eating a normal diet, glucose is the primary raw material used by cells to manufacture ATP. However there are two different methods the cell can use to generate ATP using glucose.

1. Mitochondria: A single glucose molecule can be processed by mitochondria in the presence of oxygen to produce 32 molecules of ATP.
2. Cytosol: Cellular machinery exists in the cytosol which is capable of rapidly producing 2 molecules of ATP from a single molecule of glucose without requiring oxygen as part of the process.

These two processes correspond to the "aerobic" vs "anaerobic" pathways. Running glucose through the mitochondria takes extra time, but is massively more energy efficient than processing glucose directly in the cytosol (specifically, you get 16 times more energy from a given amount of glucose if you process it through the mitochondria). But you need oxygen to process glucose through mitochondria, so if the cell runs short of oxygen (ie. you are running the 400 yard dash) processes in the cytosol will kick in to assist in the production of ATP. However, as a side effect they will also put out lactate and hydrogen ions, which change the pH of the cell and lead ultimately to inability of the muscle cell to contract, the sensation muscle burn, and huffing and puffing.

[Processing glucose through mitochondria also produces byproducts such as reative oxygen species (ROS) but we're not worrying about that right now.]

Cancer cells, however, appear to have dysfunction mitochondria, so they are unable to generate ATP by running glucose through mitochondrial processes.

This puts the cancer cell in a difficult position. All cells need a certain amount of ATP just to survive. For the sake of argument, let's say these example cells need to be able to manufacture 96 molecules of ATP per minute or else they will die. To achieve that, the non-cancerous cell simply needs to transport three molecules of glucose from the extracellular fluid and use oxygen in the mitochondria to produce 96 molecules of ATP.

On the other hand, because its mitochondria doesn't work, a cancer cell has to transport 48 molecules of glucose across the cell membrane and process all of them in the cytosol to produce the same 96 molecules of ATP that only required 3 molecules of glucose for the normal cell. In the process, the cancer cell generates a lot of lactate and hydrogen ions, which alter the pH of the environment and according to some of the articles I've read may contribute to some of the pathological traits of tumors including encouraging them to metastasize.

So, compared to normal cells, cancer cells are like mutants which burn through enormous amounts of glucose. One of the early clues that cancer was related to glucose metabolism came from patients with uncontrolled type I diabetes (prior to insulin therapy I assume). Researchers could tell when these diabetic patients got cancer because the excess glucose normally spilling out in their urine would disappear because it was being devoured by their cancer cells.

[BTW that implies that cancer cells are able to transport the glucose across the cell membrane without the action of insulin, which I think I've heard is possible, but I'm not familiar with the details of how glucose transporters and insulin work.]

With this background you can see why tumors can be detected using radioactively labeled glucose. If you imagine that each glucose molecule comes with a radioactive tag (little red stars in the figures below), here is what the normal cell would look like after producing 96 molecules of ATP.

In contrast, here is the cancer cell after making 96 molecules of ATP.

So the important question is whether or not it is possible to fight cancer by limiting the availability of the large amounts of glucose it needs to survive.